Evaluation of the practicability of limited sampling strategies for the estimation of mycophenolic acid exposure in Chinese adult renal recipients

The immunosuppressive potential of mycophenolic acid (MPA) correlates well with MPA exposure [area under the concentration-time curve (AUC)]. Monitoring MPA AUC is important and helpful for maintaining the efficacy of mycophenolate mofetil while minimizing its side effects, but full MPA AUC monitoring is laborious, cost prohibitive, and impractical. Limited sampling strategies have been proposed as an alternative method for estimating MPA exposure. The objective of this study was to evaluate the practicability of different limited sampling strategies for the estimation of MPA exposure. A total of 56 pharmacokinetic profiles from 53 adult renal recipients were used to evaluate the practicability of 10 published models. Standard correlation and linear regression analysis were used to compare the estimated MPA AUCs and corresponding full MPA AUCs, and the percentage of profiles for which prediction error fell within +/-20% was also used to assess the practicability of these models. Agreement between the estimated MPA AUCs and full MPA AUCs was further tested by Bland and Altman analysis. The model, based on four sampling time points, used the formula AUC = 12.61 + 0.37 x C-0.5 + 0.49 x C-1 + 3.22 x C-4 +8.17 X C-10,was superior to all other evaluated models, with the highest coefficient of determination (r(2) = 0.88), a low percentage prediction error (2.79%), and good agreement according to Bland and Altman analysis. Prediction errors of 87.5% (49/56) of profiles were within 20%, which was the highest of all the models. This algorithm can be reliably used for estimating MPA exposure in adult renal transplant patients treated with cyclosporine as concomitant immunosuppressant. Another model based on the formula AUC = 8.22 + 3.16 x C-0 + 0.99 x C-1 + 1.33 x C-2 + 4.18 x C-4 also has acceptable predictive performance, and it may also be practical, especially in outpatient settings, in view of its distribution of time points.